Five procedures for detecting outliers in linear regression are compared: sequential testing of the maximum studentized residual, Marasinghe's (1985) multistage procedure, and three procedures based on recursive residuals, calculated on adaptively-ordered observations. Four of these procedures test a no-outliers hypothesis. This hypothesis requires a rather stringent test criterion, meaning tha...

Amethod of generalized regression that blends tree-structured nonparametric regression and adaptive recursive partitioning with maximum likelihood estimation is studied. The function estimate is a piecewise polynomial, with the pieces determined by the terminal nodes of a binary decision tree. The decision tree is constructed by recursively partitioning the data according to the signs of the re...

We study a CUSUM–type monitoring scheme designed to sequentially detect changes in the regression parameter of an underlying linear model. The test statistic used is based on recursive residuals. Main aim of this paper is to derive the limiting extreme value distribution under the null hypothesis of structural stability. The model assumptions are flexible enough to include very general classes ...

A method of generalized regression that blends tree-structured nonparametric regression and adaptive recursive partitioning with maximum likelihood estimation is studied. The function estimate is a piecewise polynomial, with the pieces determined by the terminal nodes of a binary decision tree. The decision tree is constructed by recursively partitioning the data according to the signs of the r...

In the present study, a new correlation test based model validity monitoring procedure is proposed to online check the quality of nonlinear recursively identified models. The new method provides a simple but effective diagnose of nonlinear recursive models by detecting if the residuals are reduced to uncorrelated noise sequences. In the monitoring procedure, the correlation functions are period...

There exist two major problems in application of the conventional block BiCGSTAB method to the O(a)-improved Wilson-Dirac equation with multiple right-hand-sides: One is the deviation between the true and the recursive residuals. The other is the convergence failure observed at smaller quark masses for enlarged number of the right-hand-sides. The block BiCGGR algorithm which was recently propos...

The Fast Pole-Zero (ARMA) Recursive Least Squares Algorithm is derived using geometric projections. The new algorithm has a reduced number of computations compared to the algorithm obtained using matrix methods. In particular two inner products are eliminated. This is particularly important in multiprocessor implementations of the algorithm. The geometric approach provides insight and useful in...

Abstract Measurements of radiocarbon concentration (Δ 14 C) in fossil biogenic carbonates have been interpreted as reecting a reduced ventilation the deep Atlantic during last ice age. Here we evaluate (in)consistency an updated compilation Δ C data for deglaciation with abyssal circulation modern Atlantic. A transport equation, which mean velocity field is estimate and turbulent flux divergenc...

In this paper, ‎we propose new residuals for gamma regression models, ‎assuming that both mean and shape parameters follow regression structures. The  models are summarized and fitted by applying both classic and Bayesian methods as proposed by Cepeda-Cuervo (2001). The residuals are proposed from properties of the biparametric exponential family of distributions. ‎Simulated an...